Most Floating Car Data (FCD) systems use data of fleet management companies to monitor
the currents state of traffic. However, the original task of these systems (e.g.,
tracking positions and status of taxis or protecting vehicles) are different from
measuring travel times. Consequently, it is important to measure the ``quality'' of
a given fleet from the Floating Car Data point of view.
Usually, the number of vehicles and the mean time between two measurements is used
to assess the quality of the data but these numbers are misleading in many cases.
Other approach would be to carry out the whole process
until travel time estimation and compare the resulting travel times with other sources.
This process is very time consuming and cannot be carried out
because of lack of data in most cases.
FCD source benchmarking method -- proposed in this paper --
is a simple way of estimating, monitoring and comparison of different data sources.
The inputs of the method are the data sets only, not even the digital map is needed.
The method proved to be efficient in a Floating Car Data project in Budapest.
